Part 5: Results and Conclusion

Summary of Results

DC Philly NYC
Rows Remaining 87918.00 89104.00000 70822.00000
Rows NULL 2082.00 896.00000 19178.00000
Got through 3.00 1.00000 31.00000
Total Percent Removed 97.69 97.68778 97.72111

Before diving into this crazy results chart, I need to admit something…I don’t know what I’m talking about when it comes to GIS. If I were a real maps whiz, I would have limited my city polygons to only include areas people lived. There were issues with several areas within each city.

  1. Many of these issue areas resulted in, mostly, NULL responses from the API I used to get direction routes 9. These were easily removed.
  2. The majority of the rest were a little more slippery and got results from the API. An example of one of these were a point in the river would pull 298 NULLS but would get directions to another random point within the river. These were also easily removed (except for a small few that will be discussed below).

  3. The last bit are a bit more of a question. A question of how evil these made-up gameshow creators are. For example, several points fell into points deep within large philadelphian parks and began with 20-45 minutes of walking. These were kept in10.

    DC minutes Phi lly minutes NYC minutes DC miles Phi lly miles NYC miles
    Min 0.00000 0.00000 0.0000 0.000000 0.000000 0.000000
    1st Qu 55.30000 80.31667 83.1500 4.250792 6.364693 11.294021
    Median 72.03333 107.78333 111.7500 6.877634 10.163438 18.151150
    Mean 72.52182 116.91172 117.0508 7.119155 10.623455 19.113244
    3rd Qu 89.18333 138.43333 141.0000 9.775393 14.337647 26.475954
    Max 163.83333 516.33333 1316.0500 19.387365 41.823172 68.265572
    St. Dev. 24.61760 61.82033 79.8526 3.620629 5.549423 9.939558

So let’s start off by congratulating our nation’s capital with running away with the competition. Everything about their summary statistics tells the game show contestents to hope to see Washington DC next to their assigned coordinates.

New York City and Philadelphia are actually in pretty close in everything related to time. Unfortunatly the average milage trip in NYC was 9 miles (and 12 in the 3rd quartile) longer than in Philly! So, if you’re a real resident of NYC, you’re getting some value for your time.

The only two concerning items are the Maxs and Standard Deviations’ of New York and Philadelphia.

Visual Exploration

DC Philly NYC
Total Area 68.34 142.7 468.5
Land Area 61.05 134.1 302.6
Water Area 7.29 8.6 165.8

As suspected, we can see some Right Tail Extremes in the histograms of minutes for both NYC and Philadelphia. On the other hand, DC what a beauty. What causes this? My hypothesis is that it has to do with the size differences between the cities. You can see this in the table of Areas. This also, I think, helps explain the appearance of moreNULL values in the Philadelphia and New York routes because of the much larger amount of water area within city limits.

Talk about skew here more. Maybe even a table.

0.1 Evil Points

In order to look at these outliers within the NYC and Philly dataset, I would like to put forward an axiom:

Any trip over three hours, in this situation, can be defined as “Shit Luck”, and any trip over five hours is straight up Evil.

Below I’ve split the histograms up to show these three different types of trips. normal trips (less than 3 hours) are in light blue, bad trips are in dark blue (3 - 5 hours), and truly evil points in black (5+ hours).



Percent of Bad Trips

DC PH NY
2 hour trip 2.96 38.85 42.65
3 hour trip 0.00 8.12 5.98

What we see here is that most of the bad points are not truly evil (only 8% and 6% respectively). So, what are those points? I’ve decided to restrict these points to being above 400 minutes (you’ll see why in a second).

0.1.1 NYC Evil Trips


So we can see that all 298 trips that are above are from a singular location. Furthermore, the mean of those trips is a staggering 1227.81.

0.1.2 Philadelphia Evil Trips

Bad Origin count Bad Destination count
39.88880 288 39.88880 94
39.90100 93 39.89117 18
40.05352 33 39.97560 3
39.88409 2 40.03240 3
39.94131 2 40.05150 3



So, unlike NYC, Philadelphia has several locations that are either hard to get to or from. This is apparent from the table to the right.


## Results of Trips < 3 Hours

Let’s say, for no reason at all, that the gameshow guarantees that no trip will be more than 3 hours. How do the results change?

Sample of Short, Medium, and Long Trips

I wanted to see what some of the different length trips looked like in each city. Below is series of graphs showing short, medium, and long trips. This was done by taking sample of 50 trips from the bottom, middle, and top 10% of trips for each city.

Neighborhoods

As a last question, I was wondering where within each city was the best origin. Below are the top ten origins.



  1. I used the HERE API↩

  2. In this made-up world, I guess I’m the Evil Game Show creator. Shame for the contestents I guess….↩